Significance: Prematurity is a major health care problem in America, with more than 10,000 babies born preterm (less than 37 completed weeks of pregnancy) every week in the US. The economic cost to society of preterm birth in 2005 was more than $26 billion. Long-term outcomes of survivors often include neuro- developmental disabilities. In surviving Extremely Low Birth Weight (ELBW) babies born at less than 1000 g BW, 47% had low neurocognitive assessment (MDI score less than 70) at 18 months, indicating severe handicaps. During intensive care, these babies need many blood tests and have an immature bone marrow, thus requiring frequent blood transfusions. However, what hemoglobin levels to best transfuse at, as well as the long-term benefits and risks of Red Blood Cell (RBC) transfusion in ELBW babies, are largely unknown. Investigators and Prior Work: Only two previous trials addressed whether to transfuse at high or low hemoglobin thresholds in ELBW babies. They were done by the two clinical PIs (Kirpalani, Bell), but produced conflicting results as to whether higher hemoglobins might improve neurological outcome. If this is true, a simple therapy of RBC transfusion could be used to maintain higher hemoglobin, with potential major benefit for ELBW newborns. This question needs to be examined in a new trial, which we propose in this application. Drs. Kirpalani and Bell have joined forces with Drs Das and Brambilla, PI and Alternate PI at the Data Coordinating Center, with vast experience in clinical trials in ELBW neonates, as well as transfusion medicine. Innovation: Many neonatologists have shifted toward using lower hemoglobin thresholds for transfusion, based largely on retrospective data. We propose a randomized controlled trial (RCT) to evaluate the potential benefits of transfusing to maintain higher hemoglobin. The need to assess long-term outcomes of randomized trials is compelling in the newborn. Our primary outcome is a composite of death or neurodevelopmental impairment at 18-22 months of corrected age. Finally, the proposed study will account for differing blood bank practices and differences in dispensed products across multiple study sites and will adjust for variation in the hematocrit of dispensed RBCs by site. Approach: We will perform a RCT in ELBW babies asking: Should babies in the neonatal ICU get transfused at a higher hemoglobin threshold or a lower one, in order to have better brain development at 18-22 months of age? The question is highly relevant to the mission of the NHLBI. Trial results could radically alter blood transfusion practices in the neonatal ICU. Environment: We have partnered with the NICHD Neonatal Research Network (NRN) and its 18 clinical centers. We believe this trial is feasible and can be most efficiently conducted within the NRN centers, which have successfully completed many influential clinical trials in preterm infants that have often changed clinical practice.